Extended Target Poisson Multi-Bernoulli Filter
Yuxuan Xia, Karl Granström, Lennart Svensson, Maryam Fatemi · arXiv (Cornell University) · 2018
In this paper, a Poisson multi-Bernoulli (PMB) filter for multiple extended targets estimation is presented. The PMB filter is based on the Poisson multi-Bernoulli mixture (PMBM) conjugate prior and approximates the multi-Bernoulli mixture (MBM) in the posterior as a single multi-Bernoulli. By only having a single multi-Bernoulli representing detected targets in the posterior, the computational cost due to the data association problem can be effectively reduced. Different methods to merge the MBM are presented, along with their gamma Gaussian inverse Wishart implementations. The performance of the PMB filter is compared to the PMBM filter in different simulated scenarios.